Super Hi-Vision Terrestrial Transmission Test

2.1 Specifications of the transmitting and receiving equipment

The specifications of the transmitting and receiving equipment are listed in Table 1. The transmission frequency was 599.142857 MHz (channel 34) and the transmission output was 1 W each for the horizontal and vertical polarizations.

We used two antennas that are installed on the roof of the STRL building to separately transmit horizontal and vertical polarization signals. Although the transmitting antennas are on the same building (Figure 1), they are separated by some distance. 4L dual-loop antennas (1 stage, 2 planes) were used for both polarizations. The antenna height above the ground was 74 m.

The receiving antenna was a dual-polarization eight-element Yagi antenna (Figure 2) that can receive horizontal and vertical polarizations simultaneously. The gain of the receiving antenna was 9.2 dBd^*1 for the horizontal polarization and 10.3 dBd for the vertical polarization; the cross polarization discrimination^*2 was 20 dB or more in the forward direction. The receiving antenna height above the ground was 10 m.

2.2 Transmission parameters

The transmission parameters used in the test are listed in Table 2. The transmission frequency bandwidth was 5.57 MHz. Three carrier modulation schemes were used: 256 QAM, 1024 QAM, and 4096 QAM. The FFT size was 8 k (8,192) points, the number of carriers was 5,617, the guard interval was 126 µs, and the guard interval ratio was 1/8. The inner code was irregular LDPC^*3 with a code length of 64,800 bits and encoding ratio of 3/4. The outer code was BCH. The maximum transmission capacity for the 4096 QAM carrier modulation scheme was 39.5 Mbps for a single polarization and 78.9 Mbps for both polarizations. The horizontal and vertical polarizations were modulated with different PN signals.

2.3 Measurement locations

We selected 23 sites around STRL at which the electric field strength for the horizontal polarization would be at least 50 dBµV/m (Figure 3). The transmission point was the STRL building (center of Figure 3). The blue line in the figure indicates where the estimated horizontal polarization electric field strength is 50 dBµV/m; the red line marks the estimated electric field strength of 60 dBµV/m. The blue circles indicate the measurement points. All of the measurement points were in an urban area and were 0.9 km to 4.6 km from the transmission point.

2.4 Test system

The two modulated signals output from the modulator for the horizontal and vertical polarizations were transmitted from two antennas and received by a single dual-polarization Yagi antenna. The receiving system is shown in Figure 4. The two received signals output from the receiving antenna were passed through respective band-pass filters, attenuated by a variable attenuator, amplified by a low-noise amplifier, and then converted to IF signals by a down-converter. The two IF signals were input to a polarized MIMO-ultra-multilevel OFDM demodulator for demodulation and decoding, and the bit error rate (BER) of the bit stream for each polarization output from the demodulator was measured. At each measurement location, the received electric field strength, the BER after LDPC decoding, the required electric field strength (minimum received electric field strength for a BER of 1 x 10^-7 after LDPC decoding and a quasi-error-free (QEF) signal after BCH decoding) were measured and a line-of-sight transmission path was confirmed.

2.5 Test results

The received electric field strength measurements for the 23 locations are presented in Figure 5, where the horizontal axis indicates the measurement site and the vertical axis indicates the received electric field strength. We can see from the figure that the received electric field strength ranged from 40 dBµV/m to 80 dBµV/m and the maximum difference in received electric field strength between the polarizations was 6.5 dB. There was a line-of-sight (LOS) transmission path for 14 sites and a non-line-of-sight (NLOS) path for the other 9 sites. Also, measurement location 9 was in the shadow of a building, so the received electric field strength was very small there.

Figure 6 shows the relation between the average received electric field strength for the two polarizations and the transmission distance. Twelve of the LOS path measurement sites (indicated by solid blue circles in the figure) received QEF transmissions for both polarizations up to 4096 QAM. Two sites (solid orange triangles) received QEF transmissions for both polarizations up to 1024 QAM. Even for the NLOS paths, three sites received QEF transmissions up to 4096 QAM (open blue circles) and two sites received them up to 1024 QAM (open orange triangles). Moreover, two NLOS locations (open green squares) received QEF transmissions with 256 QAM but two others (the purple x's) did not. The solid lines in Figure 6 indicate the received electric field strength calculated under the assumption that all of the propagation loss was free-space. The average received electric field strength is fairly close to the calculated electric field strength for the 14 LOS measurement sites, but it is lower than the calculated value by at least 10 dB for the nine NLOS sites.

Table 3 lists the number of measurement sites at which QEF transmission is possible for both polarizations for each carrier modulation scheme. These results confirm that QEF transmission is possible with 4096 QAM when the received electric field strength is high, even to NLOS locations.

We obtained the required electric field strength for each carrier modulation scheme by measuring the BER after LDPC decoding for each polarization for various degrees of attenuation of the receiving antenna output for each polarization at each of the measurement sites (Table 4). For 256 QAM, the required electric field strength is averaged over 22 sites for horizontal polarization and over 21 sites for the vertical polarization; for 1024 QAM, the average is over 20 sites for both polarizations; for 4096 QAM, the average is over 19 sites for horizontal polarization and 16 sites for vertical polarization. Table 4 shows that the average required electric field strength was 46.2 dBµV/m for 256 QAM carrier modulation, 51.3 dBμµV/m for 1024 QAM, and 56.7 dBµV/m for 4096 QAM.

The above data confirm the possibility of QEF transmission of polarized MIMO-ultra-multilevel OFDM signals even in an urban propagation environment with some degree of multipath propagation.

3.1 Dual-polarization transmitting antennas

We constructed a prototype dual-polarization transmitting antenna that can transmit vertically and horizontally polarized signals simultaneously from a single antenna and installed it on the roof of the STRL building. The dual-polarization transmitting antenna used in the tests is shown in Figure 7. Inside the lightning protection cover, dipole elements for the horizontal and vertical polarizations are arranged in multiple stages to attain the same gain and directionality as the dual-loop antenna (four-loop) used at relay stations in current terrestrial digital transmission. A cross polarization discrimination of 25 dB or more was obtained within a range of about 150° in the horizontal plane.

3.2 TS splitter and combiner

For bulk transmission of a single SHV-TS signal over two channels, the SHV signal has to be split into two TS signals and the received signals have to be recombined in order to restore the original signal. We developed a TS splitter and TS combiner for this purpose.

3.3 Test system

Figure 8 shows the system for testing terrestrial Super Hi-Vision transmissions . The upper part of the figure is the transmission side of the system; the lower part is the receiving side. On the transmission side, a compressed and encoded SHV-TS signal was played by an SHV-TS player and output to a TS splitter. The TS splitter divided the SHV signal into two TS signals, which were then input to their respective polarized MIMO-ultra-multilevel OFDM modulators to produce IF signals. Next, the four modulated signals were converted to UHF channel 31 and channel 34 radio frequency (RF) signals and amplified to the specified output of 1 W. The two RF signals were then combined by the respective antenna duplexers into horizontal and vertical polarization signals and transmitted from the dual-polarization transmitting antenna installed on the roof of the STRL building.

One the receiving side, a single dual-polarization Yagi antenna received signals of both polarizations. The two received signals (RF signals) were amplified with a booster amplifier and split into the respective channels. Each polarization of each channel was input to a receiver and converted to an IF signal, resulting in a total of four IF signals. The two IF signals of each channel were demodulated and decoded by polarized MIMO-ultra-multilevel OFDM demodulators and converted into a TS signal. The TS combiner combined the two TS signals output from the demodulators into a single SHV- TS signal, which was then converted into an SHV video signal by the SHV decoder.

3.4 Test specifications

The specifications for the SHV terrestrial transmission tests are listed in Table 5. The Super Hi-Vision video signal was compressed and encoded with the MPEG-4 AVC/H.264 codec to 182 Mbps to create a data stream that would be within the transmission equipment capacity of the test system (183.6 Mbps with two channels). The signal was transmitted from a prototype multi-stage dipole dual-polarization transmitting antenna installed on the roof of the STRL building. The transmission height of the antenna was 74 m above the ground.

A dual-polarization eight-element Yagi antenna was installed on the roof of a building at the receiving site, 4.2 km north by northwest from the transmitting antenna. The signal propagation environment was an urban area with a quasi-line-of-sight path due to the presence of trees.

3.5 Test results

The electric field strength and the received signal modulation error ratio (MER)^*4 were measured at the reception site (Table 6). The electric field strength was about 60 dBµV/m for all channels and all polarizations. The MER, however, was particularly low for the vertical polarization in channel 34. The low MER was considered to be an effect of same-channel interference from a terrestrial digital transmission signal arriving at the receiving site from nearly the same direction as the test transmission. The results confirm that a stable, error-free SHV video signal can be output by the SHV decoder even in the presence of co-channel interference in the propagation environment.